We propose that change in walking speed in older individuals is induced by feelings of fatigue that occur with exercise effort and that slowing down is a compensatory strategy to reduce the level of effort and avoid fatigue. The higher the metabolic rate required for a given amount of work, the more likely an individual may approach their maximum energetic availability. Metabolic requirements for a given activity or workload derive from two sources: a. Resting metabolic rate (RMR), the energy required to maintain homeostatic equilibrium at rest and in euthermic conditions; and b. Movement efficiency, the amount of energy (required to perform one unit of work. There is emerging evidence that RMR declines with age in healthy individuals, but does not decline, or perhaps even increases, in older individuals who are sick or frail. Analogously, movement efficiency tends to decline with age, a change explained only in part by changes in body composition. Thus, with aging persons may require more energy to perform their usual activities. This increasing demand coupled with declining maximal aerobic capacity may explain why older persons perceive more fatigue than younger persons when doing the same tasks. In addition, the perception of fatigue, given the same ratio between energy availability/energy utilization may be modulated by several factors including inflammation, neurologic and other conditions related to sickness behaviour. We hypothesize that the lower fatigue threshold observed with increasing age and in frail older persons is due to: - increased RMR due to high cost of homeostasis and dysregulation or malfunction of the homeostatic network (due to both acute and chronic disease processes); - increased energetic cost of mobility due to increased biomechanical inefficiency and decline in anatomical integrity and harmonic function; - compromised energy availability, due to specific pathological processes, such as congestive heart failure, and/or inefficient transport of energy to the required sites, mainly due to reduced homeostatic ability. To test the hypotheses that the energetic cost of walking, elevated RMR and development of fatigue at low levels of activity are key predictors of disability, we propose to explore relationships among RMR, the energetic and mechanical aspects of walking using gait lab studies and the threshold for development of fatigue using participants in the BLSA. The primary focus is to evaluate whether higher RMR, high energetic cost of walking, maximum oxygen consumption during peak exercise and dysregulation of the energy homeostatic network (level of inflammatory markers and hormones) independently predict energetic threshold for the development of subjective fatigue. A secondary objective is to verify whether information on RMR, energetic cost of walking, maximum oxygen consumption during peak exercise and dysregulation of the energy homeostatic network provide information on fitness and fatigability independent of the measure of fitness provided by a standard treadmill test. Participants consist of all BLSA participants. No exclusion criteria are established a priori. Measures include oxygen consumption at rest, during normal walking and prolonged walking (400 m) using portable equipment. Oxygen consumption and level of fatigue experienced using the Borg scale will be assessed while walking at a slow pace (0.67 m/s) on the treadmill for 5 minutes and during maximal exertional threshold testing. All measures associated with energetic cost and consumption related to the energetic pathway have been implemented in the BLSA: We have introduced measures of objective and subjective fatigue using the Borg Scale and the RQ ratio, respectively. In addition, we will be measuring metabolic rate using portable equipment in different conditions, including rest, customary walking, treadmill at low load and maximum load. Part of the study is to verify the hypothesis that change in the circulating level of hormones, inflammatory markers, markers of oxidative stress and autonomic function modulates the relationship between workload, energy consumption and the development of fatigue. Finally, we have introduced in the BLSA measures of metabolic consumption during daily activity, asking participants to wear an accelerometer combined with a heart rate recorded for seven days after leaving the clinic. Using individual calibration equations elaborated during their stay in the BLSA, we plan to estimate usual daily activity and to correlate it with our measures of efficiency and energetics. A symposium highlighting key findings from this work will be given at the 2012 annual meeting of the Gerontological Society of America.